Automatic systems in which an ordinary telephone set can use a telephone line to access an automatic answering machine are known in the prior art.
Prior-art systems comprise automatic answering machines which, in response to a telephone call, permit the caller to choose from a plurality of options suggested by voice prompts in order to select predetermined paths to desired services or information.
In prior-art systems, automatic answering machines comprise programs capable of interpreting voice commands or commands activated on the telephone set via appropriate keys, as well as data or data structures which generally consist of voice prompts and are usually located at the company or organization that wishes to provide alternative service choices automatically by means of access to a single telephone number.
A disadvantage shared by all answering systems, and by automatic answering machines in particular, is the fact that both the programs used to access voice prompts and the data to be accessed reside on the same answering machine.
Because of this fact, the programs are necessarily data dependent, and the data are necessarily program dependent.
Consequently, it is not possible to develop data access procedures or programs which are independent of the type of data structure to be managed.
In addition, both programs and data are linked to the physical answering machine on which they are installed, which means that it is usually necessary to modify both programs and data if the physical answering machine is changed.
Finally, because of the close link between programs and data, any change to programs or data entails corresponding changes in the respective data or programs.
These disadvantages are even more significant in the case of answering machines that operate together with voice recognition and speech synthesis systems (voice systems).
When such configurations are used in prior-art systems, in fact, generating data-which in this case consist of synthesizable voice structures-necessarily requires that programmers be available who are capable of programming, e.g. by means of a known language such as VOICE-XML, the voice structures to be synthesized.
Essentially, the inherent structural limitations of the prior-art architecture make it impossible to set up automatic answering systems in which the procedures or programs used for accessing (instantiating) are independent of the data to be instantiated and thereby permit a greater general flexibility in implementation and access.
In addition, because of the close links between programs and data and/or the difficulties in generating multimedia structures, it is not possible with prior-art methods to implement automatic answering machines simply and without the aid of personnel with programming skills or who are otherwise highly specialized.